Predicting Critical Micelle Concentrations for Surfactants Using Graph Convolutional Neural Networks
Предсказание критических концентраций мицеллообразования поверхностно-активных веществ с использованием графовых сверточных нейронных сетей
2021-09-09
SCID: 54.1/a2h7nn6x
Discuss with AI
critical micelle concentrationgraph convolutional neural networkshigh-throughput screeningmolecular saliency mapssurfactant molecular graphs
Figures from the paper
Abstract (AI)
Surfactants are amphiphilic molecules that are widely used in consumer products, industrial processes, and biological applications. A critical property of a surfactant is the critical micelle concentration (CMC), which is the concentration at which surfactant molecules undergo cooperative self-assembly in solution. Notably, the primary method to obtain CMCs experimentally-tensiometry-is laborious and expensive. In this study, we show that graph convolutional neural networks (GCNs) can predict CMCs directly from the surfactant molecular structure. In particular, we developed a GCN architecture that encodes the surfactant structure in the form of a molecular graph and trained it using experimental CMC data. We found that the GCN can predict CMCs with higher accuracy on a more inclusive data set than previously proposed methods and that it can generalize to anionic, cationic, zwitterionic, and nonionic surfactants using a single model. Molecular saliency maps revealed how atom types and surfactant molecular substructures contribute to CMCs and found this behavior to be in agreement with physical rules that correlate constitutional and topological information to CMCs. Following such rules, we proposed a small set of new surfactants for which experimental CMCs are not available; for these molecules, CMCs predicted with our GCN exhibited similar trends to those obtained from molecular simulations. These results provide evidence that GCNs can enable high-throughput screening of surfactants with desired self-assembly characteristics.
Key Findings
1
A graph convolutional neural network predicts critical micelle concentrations directly from surfactant molecular graphs trained on experimental CMC data.
2
A single GCN model generalizes across anionic, cationic, zwitterionic, and nonionic surfactants.
3
For newly proposed surfactants lacking experimental CMCs, GCN predictions showed trends similar to molecular simulation results, supporting high-throughput screening.
4
Molecular saliency maps identify atom types and substructures influencing CMCs, consistent with established physical relationships involving constitutional and topological features.
5
The GCN achieves higher CMC prediction accuracy on a more inclusive dataset than previously proposed methods.
Research Object
Surfactant molecules and their micelle-forming self-assembly in solution
Research Subject
Prediction of critical micelle concentrations (CMCs) from molecular structure, including the effects of atom types and molecular substructures across anionic, cationic, zwitterionic, and nonionic surfactants
Publication Details
Publication Date
2021-09-09
Journal
Publisher
ISSN
Cited by
68
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest